The ability of inflammatory markers to recognize infection in cancer patients with fever at admission

Prediction of bacterial and fungal bloodstream infections using machine learning in patients undergoing chemotherapy

PURPOSE

This study aimed to develop a machine learning (ML) model to predict bloodstream infection (BSI) in chemotherapy patients.

PATIENTS AND METHODS

We included all cancer patients undergoing chemotherapy at a tertiary cancer hospital from 2017 to 2022. Data were collected per chemotherapy cycle, including chemotherapy drugs, indications, cycle number, cancer type, body mass index, age, gender, complete blood count, creatinine levels, and microbial cultures. BSI was assessed within 21 days after chemotherapy. The ML algorithms tested included logistic regression, ridge regression, k-nearest neighbors, Naive Bayes, Perceptron, neural networks, decision trees, boosting methods, Random Forests, and Support Vector Machines. The SHapley Additive exPlanations (SHAP) method was used to measure feature importance.

RESULTS

Among 107,757 cycles from 19,225 patients, 91.7 % had solid tumors, primarily breast (36.8 %) and gastrointestinal (19.4 %) cancers. The first cycle accounted for 23.7 % of cycles, and palliative chemotherapy made up 52.9 %. Alkylating agent was the most common drug class used (55.5 %). BSI occurred in 1.33 % of cycles, with 34 % of these cases occurring in neutropenic patients. Of the bacteremia cases, 11.8 % were polymicrobial, and 69.3 % involved gram-negative bacteria. The best model was a neural network with one hidden layer (5 neurons), achieving 70.7 % sensitivity, 93.49 % specificity, 93.19 % accuracy, and an area under a receiver operating characteristic curve of 91.93 %. Key predictors included the first cycle, antimetabolite use, palliative chemotherapy, monocytopenia, and hematological malignancies.

CONCLUSION

ML effectively predicts bacteremia in chemotherapy patients, including non-neutropenic cases, and could be used in clinical practice to guide treatment and infection workup.

Procalcitonin-guided antibiotic treatment in patients with cancer: a patient-level meta-analysis from randomized controlled trials

BACKGROUND

Use of serum procalcitonin (PCT), an inflammatory biomarker for bacterial infections, has shown promising results for early stopping antibiotic treatment among patients with respiratory infections and sepsis. There is need for additional data regarding effectiveness and safety of this concept among patients with cancer.

METHODS

Individual data of patients with a documented diagnosis of cancer and proven or suspected respiratory infection and/or sepsis were extracted from previous trials where adult patients were randomized to receive antibiotic treatment based on a PCT protocol or usual care (control group). The primary efficacy and safety endpoints were antibiotic exposure and 28-day all-cause mortality.

RESULTS

This individual-patient data meta-analysis included 777 patients with a diagnosis of cancer from 15 randomized-controlled trials. Regarding efficacy, there was a 18% reduction in antibiotic exposure in patients randomized to PCT-guided care compared to usual care ([days] 8.2 ± 6.6 vs. 9.8 ± 7.3; adjusted difference, - 1.77 [95% CI, - 2.74 to - 0.80]; p < 0.001). Regarding safety, there were 72 deaths in 379 patients in the PCT-guided group (19.0%) compared to 91 deaths in 398 participants in the usual care group (22.9%) resulting in an adjusted OR of 0.78 (95% CI, 0.60 to 1.02). A subgroup analysis showed a significant reduction in mortality in patients younger than 70 years (adjusted OR, 0.58 [95% CI, 0.40 to 0.86]).

CONCLUSION

Result of this individual patient meta-analysis from 15 previous trials suggests that among patients with cancer and suspected or proven respiratory infection or sepsis, use of PCT to guide antibiotic treatment decisions results in reduced antibiotic exposure with a possible reduction in mortality, particularly among younger patients.

Clinical significance of serum lactate and lactate dehydrogenase levels for disease severity and clinical outcomes in patients with colorectal cancer admitted to the intensive care unit

Objective

Serum lactate (LA) and lactate dehydrogenase (LDH) levels have a major impact on the clinical treatment of malignant tumors and critically ill patients. Nevertheless, the assessment of disease severity in oncology patients admitted to the intensive care unit (ICU) remains incomplete when considering the serum LA and LDH levels. This study aimed to investigate the significance of serum LDH and LA levels in assessing disease severity and predicting clinical outcomes in patients with colorectal cancer (CRC) admitted to the ICU.

Methods

This retrospective study included patients with CRC who were admitted to the ICU between January 2017 and December 2022. The patients were divided into three groups based on the tumor treatment methods they had received within 3 months before ICU admission: post-chemotherapy group, post-surgery group, and palliative treatment group. The association between serum LA and LDH levels and disease severity and clinical outcomes was analyzed.

Results

Of 137 patients with CRC admitted to the ICU were finally studied. Patients in the post-chemotherapy group exhibited higher serum LA and LDH levels compared to those in the other two groups. Additionally, they had higher Acute Physiology and Chronic Health Evaluation (APACHE) II scores, longer ICU length of stay, and a higher 30-day mortality. We found a significant positive correlation between serum LA levels and APACHE II scores as well as ICU length of stay and 30-day mortality. In contrast, we only observed a significant positive correlation between serum LDH levels and disease severity in the post-chemotherapy group, whereas no significant correlation between LDH levels and 30-day mortality in any of the three groups.

Conclusion

Our study concludes that elevated serum LA levels, rather than LDH levels, are more effective in assessing disease severity and could be used as predictors for clinical outcomes in patients with CRC admitted to the ICU.

Plants against cancer: the immune-boosting herbal microbiome: not of the plant, but in the plant. Basic concepts, introduction, and future resource for vaccine adjuvant discovery

The presence of microorganism communities (MOCs) comprised of bacteria, fungi, archaea, algae, protozoa, viruses, and the like, are ubiquitous in all living tissue, including plant and animal. MOCs play a significant role in establishing innate and acquired immunity, thereby influencing susceptibility and resistance to disease. This understanding has fostered substantial advancements in several fields such as agriculture, food science/safety, and the development of vaccines/adjuvants, which rely on administering inactivated-attenuated MOC pathogens. Historical evidence dating back to the 1800s, including reports by Drs Busch, Coley, and Fehleisen, suggested that acute febrile infection in response to "specific microbes" could trigger spontaneous tumor remission in humans. This discovery led to the purposeful administration of the same attenuated strains, known as "Coley's toxin," marking the onset of the first microbial (pathogen) associated molecular pattern (MAMPs or PAMPs)-based tumor immunotherapy, used clinically for over four decades. Today, these same MAMPS are consumed orally by billions of consumers around the globe, through "specific" mediums (immune boosting "herbal supplements") as carriers of highly concentrated MOCs accrued in roots, barks, hulls, sea algae, and seeds. The American Herbal Products Association (AHPA) mandates microbial reduction in botanical product processing but does not necessitate the removal of dead MAMP laden microbial debris, which we ingest. Moreover, while existing research has focused on the immune-modulating role of plant phytochemicals, the actual immune-boosting properties might instead reside solely in the plant's MOC MAMP laden biomass. This assertion is logical, considering that antigenic immune-provoking epitopes, not phytochemicals, are known to stimulate immune response. This review explores a neglected area of research regarding the immune-boosting effects of the herbal microbiome - a presence which is indirectly corroborated by various peripheral fields of study and poses a fundamental question: Given that food safety focuses on the elimination of harmful pathogens and crop science acknowledges the existence of plant microbiomes, what precisely are the immune effects of ingesting MAMPs of diverse structural composition and concentration, and where are these distributed in our botanicals? We will discuss the topic of concentrated edible MAMPs as acid and thermally stable motifs found in specific herbs and how these would activate cognate pattern recognition receptors (PPRs) in the upper gut-associated lymphoid tissue (GALT), including Peyer's patches and the lamina propria, to boost antibody titers, CD8+ and CD4+ T cells, NK activity, hematopoiesis, and facilitating M2 to M1 macrophage phenotype transition in a similar manner as vaccines. This new knowledge could pave the way for developing bioreactor-grown/heat-inactivated MOC therapies to boost human immunity against infections and improve tumor surveillance.